Before a sensitive tool call runs, ask Vulnify and get ALLOW, REVIEW or BLOCK,
with a 0–100 risk score and the reasons behind it.
Website · Docs · MCP server · X
How it works
- Your agent's code calls
check()orguard()before it reads, writes, deletes, exports or sends data. - REVIEW pauses only that action until a different person approves it, in Vulnify or in Slack.
- Start in monitor mode. Checks fail closed by default.
- Policies are code: pull, test and apply them from the CLI.
- Every decision lands in a hash-chained audit log, exportable as JSON or CSV.
Open source
| Repo | What it is |
|---|---|
| vulnify-sdk | Node.js SDK and CLI · npm i @vulnify/sdk · LangChain, Vercel AI SDK, OpenAI Agents SDK and MCP adapters |
| vulnify-sdk-python | Python SDK and CLI · pip install vulnify · CrewAI and LangGraph adapters |
| vulnify-mcp | MCP server for Claude, Cursor and Grok · hosted at https://mcp.vulnify.io/mcp with OAuth |
Get started: free plan · Security: security@vulnify.io